summaryrefslogtreecommitdiffstats
path: root/health/health.d/hdfs.conf
diff options
context:
space:
mode:
Diffstat (limited to '')
-rw-r--r--health/health.d/hdfs.conf75
1 files changed, 75 insertions, 0 deletions
diff --git a/health/health.d/hdfs.conf b/health/health.d/hdfs.conf
new file mode 100644
index 0000000..678faab
--- /dev/null
+++ b/health/health.d/hdfs.conf
@@ -0,0 +1,75 @@
+
+# make sure hdfs is running
+
+template: hdfs_last_collected_secs
+ on: hdfs.heap_memory
+ calc: $now - $last_collected_t
+ units: seconds ago
+ every: 10s
+ warn: $this > (($status >= $WARNING) ? ($update_every) : ( 5 * $update_every))
+ crit: $this > (($status == $CRITICAL) ? ($update_every) : (60 * $update_every))
+ delay: down 5m multiplier 1.5 max 1h
+ info: number of seconds since the last successful data collection
+ to: webmaster
+
+
+# Common
+
+template: hdfs_capacity_usage
+ on: hdfs.capacity
+ calc: ($used) * 100 / ($used + $remaining)
+ units: %
+ every: 10s
+ warn: $this > (($status >= $WARNING) ? (70) : (80))
+ crit: $this > (($status == $CRITICAL) ? (80) : (98))
+ delay: down 15m multiplier 1.5 max 1h
+ info: used capacity
+ to: sysadmin
+
+
+# NameNode
+
+template: hdfs_missing_blocks
+ on: hdfs.blocks
+ calc: $missing
+ units: missing blocks
+ every: 10s
+ warn: $this > 0
+ delay: down 15m multiplier 1.5 max 1h
+ info: missing blocks
+ to: sysadmin
+
+
+template: hdfs_stale_nodes
+ on: hdfs.data_nodes
+ calc: $stale
+ units: dead nodes
+ every: 10s
+ warn: $this > 0
+ delay: down 15m multiplier 1.5 max 1h
+ info: stale data nodes
+ to: sysadmin
+
+
+template: hdfs_dead_nodes
+ on: hdfs.data_nodes
+ calc: $dead
+ units: dead nodes
+ every: 10s
+ crit: $this > 0
+ delay: down 15m multiplier 1.5 max 1h
+ info: dead data nodes
+ to: sysadmin
+
+
+# DataNode
+
+template: hdfs_num_failed_volumes
+ on: hdfs.num_failed_volumes
+ calc: $fsds_num_failed_volumes
+ units: failed volumes
+ every: 10s
+ warn: $this > 0
+ delay: down 15m multiplier 1.5 max 1h
+ info: failed volumes
+ to: sysadmin